.=) 1991 Oxford University Press
Nucleic Acids Research, Vol. 19, No. 9 2295
Physical map of the centromeric region of human
chromosome 7: relationship between two distinct alpha
satellite arrays
Rachel
Wevrick1l2 and Huntington F.Willard' *
1Department of Genetics, Stanford University, Stanford, CA 94305, USA and 2Department of
Molecular and Medical Genetics, University of Toronto, Toronto, Ontario M5S 1A8, Canada
Received February 19, 1991; Revised and Accepted April 9, 1991
ABSTRACT
A long-range physical map of the centromeric region
of human chromosome 7 has been constructed in order
to define the region containing sequences with
potential involvement in centromere function. The map
is centered around alpha satellite DNA, a family of
tandemly repeated DNA forming arrays of hundreds to
thousands of kilobasepairs at the primary constriction
of every human chromosome. Two distinct alpha
satellite arrays (the loci D7Z1 and D7Z2) have
previously been localized to chromosome 7. Detailed
one- and two- locus maps of the chromosome 7
centromere have been constructed. Our data indicate
that D7Z1 and D7Z2 arrays are not interspersed with
each other but are both present on a common Mlu I
restriction fragment estimated to be 3500 kb and 5500
kb on two different chromosome 7's investigated.
These long-range maps, combined with previous
measurements of the D7Z1 and D7Z2 array lengths, are
used to construct a consensus map of the centromere
of chromosome 7. The analysis used to construct the
map provides, by extension, a framework for analysis
of the structure of DNA in the centromeric regions of
other human and mammalian chromosomes.
INTRODUCTION
The elucidation of the DNA sequences which comprise the
centromeres of eukaryotic chromosomes remains a key step in
the understanding of how this chromosomal element functions
in the attachment of chromosomes to the mitotic and meiotic
spindles (1, 2). The minimum requirement for centromeric DNA
is that it contain sequences which form a recognition site for
kinetochore proteins (3). The region in which such sequences
exist is defined functionally only in the lower eukaryotes
Saccharomyces cerevisiae and Schizosaccharomyces pombe, and
the exact sequences are known only in S. cerevisiae (4). In
mammalian chromosomes, the centromere is defined
cytogenetically by the primary constriction of a metaphase
*
To whom correspondence should be addressed
chromosome. Given current cytogenetic resolution, this region
may encompass many millions of basepairs in each chromosome.
At present, the best candidate for a sequence important to
centromere structure in human chromosomes is alpha satellite
DNA, a family of tandemly repeated DNA present at every
human centromere (5, 6, 7). A specific sequence from some alpha
satellite repeats (8) has been shown to bind a centromere-specific
protein, CENP-B (9), which has been independently localized
to the heterochromatin subjacent to the kinetochore (10). The
demonstrated relationship between alpha satellite DNA and
CENP-B (8, 10, 11), and the localization of alpha satellite at every
human centromere suggests that alpha satellite DNA may play
a role in the structure or function of the human centromere (5,
11). Other candidate centromere sequences may be difficult to
recognize in the absence of a functional assay. In addition, a better
understanding of the physical organization of the sequences which
are present in the region of the primary constriction is required.
Alpha satellite DNA is a primate-specific satellite DNA family
composed of 171 bp monomer units, which are tandemly
arranged in a largely chromosome-specific manner to form
higher-order repeat units (6). Both the length of alpha satellite
arrays and the distribution of particular restriction sites in and
flanking the arrays are highly variable between homologous
human chromosomes (12, 13, 14). Despite this extensive
polymorphism, it has been possible to obtain an estimate of the
size of arrays on different homologous and non-homologous
chromosomes (12, 14, 15, 16).
The centromeric region of chromosome 7 contains two distinct
arrays of alpha satellite (17, 18). D7Z1 is composed of multiple
six monomer higher-order repeat units, tandemly repeated to form
an array of average length -2580 kb, with a range on different
copies of chromosome 7 of 1530-3810 kb, estimated using
pulsed-field gel electrophoresis (12). D7Z2 is one of the smallest
arrays measured, at -265 kb (range 100-550 kb). In this work,
we present a long-range physical map of the centromeric region
of chromosome 7, including both alpha satellite arrays, for two
different chromosome 7's isolated in rodent-human somatic cell
hybrids.
-
2296 Nucleic Acids Research, Vol. 19, No. 9
MATERIALS AND METHODS
Somatic cell hybrid DNA
A50-1Ac13A is a mouse-human somatic cell hybrid containing
as its only human component a translocation between the X
chromosome and chromosome 7, which carries one chromosome
7 centromere and no other human centromere (18). KO15 is a
hamster-human somatic cell hybrid containing a single
chromosome 7 as its only human component (19). Cytogenetically
the chromosome 7 appears to have a short arm rearrangement
not involving the centromere (unpublished data). DNA was
isolated for pulsed-field gel electrophoresis as previously
described (11).
In situ hybridization
Metaphase chromosomes from A50-lAcl3A and a normal male
were prepared and hybridized to biotinylated probes pMGB7
(detecting the locus D7Z2) or pa7tl (detecting the locus D7Z1)
essentially as previously described (16). Biotinylated probes were
obtained from Oncor, Inc. (Gaithersburg, MD.). High stringency
conditions included hybridization at 370C in 65% formamide,and
washing at 430C in 65% formamide. Detection of the
hybridization signal
was
performed using a commercially
available chromosome in situ kit (Oncor, Inc.). Slides were
counterstained in propidium iodide and destained before viewing.
Pulsed-field gel electrophoresis and Southern hybridization
Restriction endonuclease digestion, electrophoresis, transfer to
nylon membrane, prehybridization and hybridization were
performed essentially as described (20). Hybridization of
Southern blots with each probe was performed under conditions
of high stringency (530C hybridization in 50% formamide,
3 x SSC, final wash in 0.1 xSSC at 65°C). Contour-clamped,
homogeneous electric field electrophoresis (CHEF) was
performed using the CHEF-DRII system (BioRad). Conditions
for the CHEF gel electrophoresis are given in the figure legends.
Yeast chromosomes (strain YNN295, Bio-Rad), lambda DNA
concatemers (Pharmacia) and lambda DNA digested with Hind
Ill were used as size markers.
Fig. 1 In situ hybridization of the biotin-labeled probes pMGB7 (locus D7Z2) and pot7tl (locus D7Zl) to metaphase chromosomes from a somatic cell hybrid and
a normal human male. A) probe pMGB7, cell line A50-lAc13A, B) pMGB7, normal male, C) pca7tl, A50-lAc13A, D) pa7tl, normal male. In each case, the
signal is localized to the primary constriction of the chromosome 7 (KO15) or both chromosome 7 homologues (normal male).
Nucleic Acids Research, Vol. 19, No. 9 2297
RESULTS
Localization of D7Z1 and D7Z2 by in situ hybridization
The loci D7Z1 and D7Z2 were originally assigned to human
chromosome 7 by Southern blot hybridization of DNA from a
hybrid panel (18). In order to establish the presumed centromeric
localization of the two arrays, in situ hybridization of the probes
pMGB7 and pa7tl to metaphase chromosomes was performed.
Using conditions of high stringency under which the two probes
do not cross-hybridize (12, 18), biotinylated pMGB7 hybridizes
to the centromere of the 7/X translocation chromosome in
A50-lAc13A (Fig. 1, panel A) and to the centromere of the two
chromosomes 7 in a normal male (Fig. 1, panel B). Similarly,
using biotinylated pa7tl as a probe under high stringency
conditions, the chromosome 7 centromeres in both A50-lAc13A
and the human cells are specifically labeled (Fig. 1, panels C
and D). The more intense signal from the pa7tl probe reflects
the larger target size of the D7Z1 array, which is on average
ten times larger than the D7Z2 array (12). The signals from the
two probes appeared to be coincident at the primary constriction
at this level of resolution.
Pulsed-field mapping of two alpha satellite arrays: single-locus
maps
We turned to pulsed field gel electrophoresis in order to define
the physical relationships between the two alpha satellite arrays.
In previous experiments using the two alpha satellite probes
pMGB7 and pa7tl to detect large alpha satellite DNA-containing
restriction fragments on CHEF blots of DNA from individuals
in a large three-generation family, it had been demonstrated that
both the array sizes and the placement of individual restriction
sites within and outside of the arrays varies between copies of
chromosome 7 (12). Thus the analysis of single chromosome 7's
isolated in two different somatic cell hybrids was necessary.
Knowledge of the restriction map of single higher-order repeats
belonging to each chromosome 7 array (18) allowed a choice
of useful restriction enzymes for long-range mapping(12). DNA
from each of two somatic-cell hybrids containing a chromosome
7 centromeric region was digested with a set of 17 restriction
enzymes, expected to have few or no sites within each array,
and analysed by CHEF gel electrophoresis and Southern blotting.
If restriction sites for the particular enzyme existed within the
tandem array, several fragments hybridized to the alpha satellite
probe since each probe detects a locus spanning hundreds to
thousands of kb of DNA. Using single and double digests for
this class of enzyme (e.g. BamH I and Bgl II for D7Z2 in
A50-lAcl3A, lanes 2 and 3, Fig. 2A) a restriction map was
developed for the internal sites of each of the two arrays in each
chromosome 7. These fragment sizes were tallied and their sum
gives a maximum estimate of the array size. This sum was
consistent between these enzymes, and these enzyme sites are
virtually coincident at the edges of the arrays (Fig. 3). For
example, using this approach the locus D7Z2 was estimated to
span 170 kb in A50-lAcl3A, determined with restriction enzymes
such as BamH I, Bgl II, and Xba I.
A second class of enzymes revealed only a single fragment
on a Southern blot of a CHEF gel, either of the same size as
estimated above (e.g. Eco RV for the D7Z2 in A50-1Acl3A,
lane 5, Fig. 2A) or a larger size (e.g. Apa I and BstE II, lanes
1 and 4, Fig. 2A). These enzymes were used to develop a map
of the sites flanking each array. If double digests with these
enzymes failed to yield a smaller fragment than the smaller of
the two fragments detected in either of the single enzyme digests,
then it was possible to determine the order of these flanking sites,
but not their distances from the ends of the array. The blot used
for pMGB7 in 2A, lanes 1-5 was stripped of this probe and
rehybridized under high stringency to pa7tl, revealing a signal
in the compression zone of the gel in the Apa I and BamH I
digests (lanes 6 and 7). No cross-hybridization to the fragments
from lanes 1 and 2 was detected.
Each probe was used in this fashion to develop a map in and
around the array to which the probe hybridizes, independent of
the second probe (Fig. 3A, D7Z1 and 3B, D7Z2). The maps
A
D7Z2
D7Z1
I
.
_
E
<
co
m
m
u
<
1.
2.
3.
4.
5.
6.
7.
-I
I
L
size (kb)
CD)
U)
UJ
L
u)
c)
co
CL
E
U
co
>950 700 -
290-
_
23-
B
D7Z2
D7Z1
ut
U)
cn
5i
co
size (kb)
>6000
-
1.
0
2.
=
-
U
z
c)
(I)
U,in
3.
4.
5.
6.
.*.
I
_
U)
-I
c-0
co
z
7.
o.
8.
9.
-
D
=
E
Z
cX u cn
10. 11. 12.
A&.I
W.,
4700 .;K.W
3500 2200 -
.11
_
&.
d- a
245 -
Fig. 2 CHEF gel restriction analysis of the alpha satellite arrays on chromosome
7 (A, hybrid A50-IAcl3A, B, hybrid K015). The restriction enzymes used for
DNA digestion are marked above each lane, and size markers, measured by
YNN295 yeast chromosomes (Bio-Rad) and lambda DNA digested with Hind
III, are indicated on the left. In addition, lambda ladders were used as markers
to size accurately fragments between 23 and 290 kb in size. The locus detected
in each case is noted by the D number above the lanes. Conditions for the gel
in 2A were 1 % agarose in 0.5 xTBE, 200V for 24 hours, with a pulse time ramped
from 20-80s. Conditions for the gel in 2B were 0.7% Megarose (Clontech) in
0.5 xTBE, 50V, 120h at a pulse time of one hour followed by 20h at a pulse
time of 90s.
2298 Nucleic Acids Research, Vol. 19, No. 9
A
>
Brn
Bg
Sf
Sm
EV
P
II
II
II
613
P
el
EV
II
II
Ig
EV
P
I
I
Sm
I
St
I
0
D7ZI
A-Apa I
Bc-Bcl I
Og-Bgl 11
BI-Bgl I
Bm-BamH I
Bs-BstE II
El-EcoR I
EV-EcoR V
P-Pvu 11
st-sfi I
Sm-Sma I
X-Xba I
A50-1Acl3A
as
SC
Bs
Bc
Bg
Og
St
Sm
I
p
H
Bm
9P
I Il
I
H
P B
H
Il
II
p
H
H Sm
lI
Sf
Sm
I
D7ZI
SCALE
l
2
K01 5
250 kb
B
19
am
EV
p
?I
f1 A B1X
II I I
Sm
am
ag
I
Sg
Sm
EV
p
X X B8 A
I I1 I
9
Sm
I
Sf
D7Z2
A50-1 Ac13A
El
EV X
Sf
Sm Bl
II
Bs
!LBi'g XBg El
I
11111 1
SCALE
B,
Sm
El BSc
X PEV Bs
El
I
I I
l
el Sm
St
I I
D7Z2
K01 5
50 kb
Fig. 3 Single-locus restriction maps were derived for each locus separately, using the restriction enzymes listed in the key. The rectangles represent the regions
containing alpha satellite DNA. Not all enzymes were appropriate for both maps. Note the difference in scale between the maps in A (probe pa7tl, locus D7ZI)
and those in B (probe pMGB7, locus D7Z2). The name of the somatic cell hybrid carrying the chromosome 7 from which the map was derived is listed below
the name of the locus.In some cases, the restriction sites were placed in a symmetric arrangement because it was not possible to derive the distances from each
edge of the array, only the total distance between sites. See the text for further information concerning the placement of specific restriction sites.
confirm previous results from partial digests of A50-lAcl3A
DNA on conventional gels which suggested that D7Z1 and D7Z2
were separate arrays, rather than being interspersed with each
other (18). Because the relative distances between the flanking
sites on either side of each array could not be determined, these
sites are placed arbitrarily in equidistant positions on either side
of the array. This gives the flanking sites of the single locus maps
a symmetrical appearance which may not be reflective of the
situation on the chromosome, but indicates only the order of these
sites (Fig. 3). In addition, because of the much larger size of
the D7Z1 array and the decreased resolution of CHEF gels in
this size range, the potential error in the position of the flanking
sites for this array is larger than that for the smaller D7Z2 array.
Pulsed-field mapping of two alpha satellite arrays: two-locus
maps
Standard rare-cutting enzymes (i.e., those with CpG-rich
recognition sites) were used next to define the position of the
arrays with respect to each other. Restriction sites for some such
enzymes exist between the two arrays. For example, D7Z2 is
present on a 450 kb Sma I fragment, and D7Z1 on a 1500 kb
Sma I fragment in KO15 (Fig. 2B, lanes 5 and 11). Partial digests
using either Sfi I or Sma I gave a result consistent with the D7Z2
and D7Z1 arrays residing on adjacent Sfi I restriction fragments,
and adjacent Sma I fragments (upper bands, in lanes 4,5,10 and
11, Fig. 2B). Finally, a limited number of rare-cutting enzymes
gave the same size fragment with either probe, indicating that
the two arrays could be co-localized on the same large restriction
fragment (e.g. Mlu I and Sst II digest of K015, Fig. 2B, lanes
2,6,8, and 12). Both the Mlu I and Sst II digests are partial, as
seen by the presence of the upper band in each lane.
Using combinations of the types of enzymes listed above,
physical maps were constructed of the region containing both
D7Z1 and D7Z2 for both chromosome 7's examined (Fig. 4A).
These two-locus maps could be constructed using the information
provided by restriction fragments which harboured both arrays.
However, the orientation of the initial one-locus maps with respect
to the two-probe map, as well as the orientation of the two-locus
Nucleic Acids Research, Vol. 19, No. 9 2299
A
Sf
Sm
St
Sm
Ss
St
Sm
Ss
lm
Oh
II
I.. I I
I
N M
Ssl
Bh
M N
Sm am
III~~~~~~~~~~~~~~~
II
I I I Iwlw
am
IA50-1 AcI3A
I
D7Z2
D7Z1
Sf
Sm
Sf
Sm
Bh-BssH 11
Bm-BamH I
M-Mlu I
N-Not I
sf-Sfi I
Sm-Sma I
Oh
N
II-Ia.
MI
Ss
Sm
I
1
vw~~
St
Sm
Sm
Sm
I
~
I
~~~~~~~~~
z
.1.
I
St
I
m
SJ
Oh
N
I
K01 5
D7Z2
D7ZI
Ss-Sst 11
Sm
SCALE
HH
250 kb
B
Oh
N
M
M
I
toII
I
L-1
J
,,
4-
7q
D7Z1
(1500-3800 kb)
???
<X M
-
D7Z2
(100-550 kb)
Bh
N
-
nof I
%yn -
Consensus map
7p
Fig. 4. A. Using the individual maps presented in Fig. 3 and the additional enzymes listed in the key, two-locus restriction maps were constructed which include
both alpha satellite arrays on each chromosome. The wavy lines indicate breaks in the map, for cases in which the restriction fragment produced by digestion at
sites outside the breaks were not resolved. As described in the text, the bracket below a set of restriction enzyme sites indicates that the site could not be placed
exactly although the size of the restriction fragment produced by that enzyme is known. B. A consensus map of the centromeric region of chromosome 7 (discussed
in the text) was derived from the two maps presented in A and previous data on the sizes of chromosome 7 arrays (12), with the shaded rectangles representing
the alpha satellite arrays. The range in size of each array is indicated below the locus name. Certain rare-cutting enzymes (e.g. Not I, Mlu I and BssH II) are expected
to release a fragment containing both arrays in most (or all) chromosome 7's.
map with respect to the chromosome arms remain to be
determined (see Discussion). The exact distance between the two
arrays on each copy of chromosome 7 examined cannot be
measured accurately. However, it can be estimated to be less
than 1000 kb based on the difference between the size of each
Sfi I (or Sma I) fragment and the BamH I fragment (which
approximates the array size) within it (Fig. 4A). Two enzymes,
Not I and BssH II, gave unresolved bands of greater than 6 Mb
upon restriction enzyme analysis in all experiments. These sites
were therefore placed at the boundaries of the two-locus maps,
with breaks in the maps to indicate that the sizes of these
fragments are unknown.
In the two-locus maps, brackets under the restriction enzyme
sites between the two arrays reflect uncertainty in the order of
the sites, which cannot be ordered unambiguously without the
use of a third probe from this region. Thus, although the exact
sizes of the restriction fragments produced by these enzymes are
known, the placement of each site can be indicated only
approximately in Fig. 4A. Although the size of the brackets
indicates the region in which these particular sites are located,
the single-locus maps (Fig. 3) give more direct information in
this regard.
-
DISCUSSION
The isolation of the DNA sequences required for structural
elements of chromosomes, such as centromeres, requires a
fundamental understanding of the nature and organization of
sequences in the region of interest. In the centromeric region,
alpha satellite DNA makes up the bulk of the DNA so far
identified (5) and has thus formed the basis for our long-range
restriction maps. In both chromosomes examined, the two arrays
are present on a single large rare-cutting fragment of greater than
3.5 Megabases (Fig. 2). Given the localization of alpha satellite
DNA at the primary constriction of chromosome 7 by in situ
hybridization, it is not unreasonable to expect that the maps
presented here contain some if not all of the sequences responsible
for centromere activity.
Because of the extensive polymorphism in restriction sites and
alpha satellite array lengths between chromosome 7 homologues
(12), the maps presented are accurate in detail only for the two
copies of chromosome 7 described. However, the general features
of the maps are representative of the chromosome 7 centromere
region and some features may extend to other centromeres as
well. In each case, the restriction maps of the D7Z1 and D7Z2
arrays and surrounding DNA could be developed independently,
implying that the two arrays are not interspersed and are separated
by less than 1 Mb. The sizes of the D7Z1 arrays (3100 kb and
1800 kb) and the D7Z2 arrays (170 kb and 230 kb) are within
the range previously measured (12). Although the orientation of
the maps with respect to the rest of the chromosome is not known
with certainty, the D7Z1 array can be tentatively assigned to the
long arm side of the centromeric region, and D7Z2 to the short
arm side by two different lines of evidence. An isochromosome
2300 Nucleic Acids Research, Vol. 19, No. 9
for the short arm of chromosome 7 identified in a glioma cell
line lacked D7Zl sequences as detected by in situ hybridization
(21),thus assigning D7Z1 to the missing long arm portion of this
rearrangement. This orientation has been confirmed in two
independent studies by simultaneous two colour in situ
hybridization of the two probes (D. Pinkel, personal
communication, and T. Haaf, personal communication).
It was a priori possible that there were multiple copies of each
array with identical restriction maps present on each chromosome
7. However, this can now be ruled out since there is insufficient
space in the Sma I fragments which contain the individual arrays
to accomodate two copies of the array, and there is insufficient
space in the Mlu I fragments which contain the pair of arrays
to permit a tandem duplication.
A 'consensus' centromere map (Fig. 4B), a generic version
of these two centromere maps, was constructed, using the maps
in Fig 4A and the previously determined range in size of the two
arrays on chromosome 7 (12). In this map, the flexibility in array
sizes is indicated by the range in size below the shaded box which
represents the average size alpha satellite array. The distance
between the two arrays is approximated as up to 1 Mb from the
data in Fig. 4A. Although much information is lost in the transfer
between the map of this region in a particular chromosome and
a consensus map, the consensus map provides information
concerning the expected appearance of this region in any
chromosome 7, including array size estimates and expected
restriction sites. The rare-cutting restriction fragments adjacent
to the one containing both alpha satellite arrays are expected to
contain either typical single copy sequence DNA from the
chromosome arms or a third type of repetitive DNA. If they
contain typical sequence DNA, then it can be cloned and mapped
using standard methods. This DNA is then likely to lie on rarecutting restriction fragments which are not as highly polymorphic
as are the alpha satellite-containing rare-cutting fragments.
If the adjacent rare-cutting restriction fragment contains DNA
of non-typical sequence, then a separate consensus map of this
DNA must be constructed and joined to the alpha satellite-based
maps presented here. The need for additional consensus maps
is predicted for centromeric and pericentromeric physical maps
on chromosomes which are known to contain classical satellite
DNA (22) (e.g. chromosomes 1, 9, 16 and Y) or beta satellite
DNA (23) (e.g. at least chromosomes 1, 9, and the acrocentrics).
In either case, the development of consensus maps for regions
of repetitive, highly polymorphic DNA is essential to the
completion of physical chromosome maps (24).
The data presented in Figures 2-4 demonstrate that the D7Z1
and D7Z2 arrays are separated by DNA of unknown sequence.
The appearance of restriction sites between the two arrays for
enzymes (such as Sma I and Sfi I) which do not ordinarily have
sites in any alpha satellite DNA (25) may indicate at least some
of this DNA is not typical alpha satellite DNA. Some of this DNA
may consist of alpha satellite which is diverged in sequence from
either of the two probes, and so is not detected by them at high
stringency. Identification of this DNA awaits the cloning of
sequences adjacent to the arrays. In addition, the presence within
the arrays of restriction sites for enzymes which do not ordinarily
cut within the higher-order repeat unit was noted. However, the
sites are not clustered as might be expected if a foreign sequence
were embedded in the array, implying that the presence of these
sites may merely reflect the 1-2% differences in sequence
between higher-order repeat units (6, 26).
Experience in mapping in centromeric regions indicates that
-
cloning of the DNA across the centromeric region may also
require a different approach from that used for the chromosome
arms. The advent of yeast artificial chromosome (YAC)
techniques (27) has now made it possible to isolate larger
fragments of DNA, with a greater likelihood of detecting the
junction between alpha satellite DNA and flanking sequences.
However, because YACs containing tandem alpha satellite repeats
have been shown to be prone to rearrangement (28), YACs may
not represent a faithful copy of the original genomic structure.
As well, restriction site biases intrinsic to each alpha satellite array
may preclude the use of some YAC libraries with small insert
sizes for arrays which do not contain certain restriction sites for
tens or hundreds of kb of DNA. Despite these drawbacks, YAC
technology can still be used for mapping centromeric DNA (28).
Single- or low-copy probes isolated from YACs or cosmids
containing repetitive DNA can be added to the consensus map
using pulsed-field gel mapping.
We have developed a physical map of the centromere of
chromosome 7 as a guide for the identification and cloning of
elements responsible for centromere activity. The construction
of consensus maps of regions of the genome, including
centromeres, which may be difficult to map and clone due to
extensive variation between different copies of the same region,
is a necessary prelude to identifying regions of potential
significance. Finally, the ability to clone such sequences into YACs or other suitable vectors using a strategy guided by physical
maps of the region may eventually allow the development of a
functional assay for centromere sequences in mammalian cells.
ACKNOWLEDGEMENTS
We thank Peter Warburton, Gillian Greig and Lap-Chee Tsui
for helpful discussions, and Vicki Powers for performing the in
situ hybridization experiments. This work was supported by
research grant HGO0107 from the National Institutes of Health
and research grant # 1-1178 from the March of Dimes Birth
Defects Foundation. R.W. was supported by a studentship from
the Medical Research Council of Canada.
REFERENCES
Brinkley, B.R. (1990) Cell Motil. Cytoskeleton, 16, 104-109.
Rieder, C.L. (1982) Int. Rev. Cytol., 79, 1-58.
Nicklas, R.B. (1971) Adv. Cell Biol., 2, 225-298.
Clarke, L. (1990) Trends Genet., 6, 150-154.
Willard, H.F. (1990) Trends Genet., 6, 410-416.
Willard, H.F. and Waye, J.S. (1987) Trends Genet., 3, 192-198.
Manuelidis, L. (1978) Chromosoma, 66, 23-32.
Masumoto, H., Masukata, H., Muro, Y., Nozaki, N. and Okazaki, T. (1989)
J. Cell Biol., 109, 1963-1973.
9. Earnshaw, W.C., Sullivan, K.F., Machlin, P.S., Cooke, C.A., Kaiser, D.A.,
Pollard, T.D., Rothfield, N.F. and Cleveland, D.W. (1987) J. Cell Biol.,
104, 817-829.
10. Cooke, C.A., Bernat, R.L. and Earnshaw, W.C. (1990) J. Cell Biol., 110,
1.
2.
3.
4.
5.
6.
7.
8.
1475-1488.
11. Wevrick, R., Eamshaw, W.C., Howard-Peebles, P.N. and Willard, H.F.
(1990) Mol. Cell. Biol., 10, 6374-6380.
12. Wevrick, R. and Willard, H.F. (1989) Proc. Natl. Acad. Sci. (USA), 86,
9394-9398.
13. Jabs, E.W., Goble, C.A. and Cutting. G.R. (1989) Proc. Natl. Acad. Sci.
USA, 86, 202-206.
14. Oakey, R. and Tyler-Smith, C. (1990) Genomics, 7, 325-330.
15. Mahtani, M.M. and Willard, H.F. (1990) Genomics, 7, 607-613.
16. Greig, G.M., England, S.B., Bedford, H.M. and Willard, H.F. (1989) Am.
J. Hum. Genet, 45, 862-872.
17. J0rgensen, A.L., Bostock, C.J. and Bak, A.L. (1986) J. Mol. Biol., 187,
185-196.
Nucleic Acids Research, Vol. 19, No. 9 2301
18. Waye, J.S., England, S.B. and Willard, H.F. (1987) Mol. Cell. Biol., 7,
349-356.
19. Rommens, J.M., Zengerling, S., Bums, J., Melmer, G., Kerem, B., Plavsic,
N., Zsiga, M., Kennedy, D., Markiewicz, D., Rozmahel, R., Riordan, J.R.,
Buchwald, M. and Tsui, L.-C. (1988) Am. J. Hum. Genet., 43, 645-663.
20. Willard, H.F., Smith, K.D. and Sutherland, J. (1983) Nuc. Acids Res., 11,
2017-2033.
21. Cremer, T., Lichter, P., Borden, J., Ward, D.C. and Manuelidis, L. (1988)
Hum. Genet., 80, 235-246.
22. Prosser, J., Frommer, M., Paul, C. and Vincent, P.C. (1985) J. Mol. Biol.,
187, 145-155.
23. Waye, J.S. and Willard, H.F. (1989) Proc. Natl. Acad. Sci. (USA), 86,
6250-6254.
24. Tyler-Smith, C. (1987) Development, 101, 93-100.
25. Waye, J.S. and Willard, H.F. (1987) Nucleic Acids Res., 15, 7549-7569.
26. Durfy, S.J. and Willard, H.F. (1989) Genomics, 5, 810-821.
27. Burke, D.T., Carle, G.F. and Olson, M.V. (1987) Science, 236, 806-812.
28. Neil, D.L., Villasante, A., Fisher, R.B., Vetrie, D., Cox, B. and TylerSmith, C. (1990) Nucleic Acids Res., 18, 1421-1428.